This paper presents the JGU Mainz submission to the WMT25 Shared Task on LLMs with Limited Resources for Slavic Languages: Machine Translation and Question Answering, focusing on Ukrainian, Upper Sorbian, and Lower Sorbian. For each language, we jointly fine-tune a Qwen2.5-3B-Instruct model for both tasks with parameter-efficient finetuning. Our pipeline integrates additional translation and multiple-choice question answering (QA) data. For Ukrainian QA, we further use retrieval-augmented generation. We also apply ensembling for QA in Upper and Lower Sorbian. Experiments show that our models outperform the baseline on both tasks.

Diagrams convey symbolic information in a visual format rather than a linear stream of words, making them especially challenging for AI models to process. While recent evaluations suggest that vision-language models (VLMs) perform well on diagram-related benchmarks, their reliance on knowledge, reasoning, or modality shortcuts raises concerns about whether they genuinely understand and reason over diagrams. To address this gap, we introduce Chimera, a comprehensive test suite comprising 7,500 high-quality diagrams sourced from Wikipedia; each diagram is annotated with its symbolic content represented by semantic triples along with multi-level questions designed to assess four fundamental aspects of diagram comprehension: entity recognition, relation understanding, knowledge grounding, and visual reasoning. We use Chimera to measure the presence of three types of shortcuts in visual question answering: (1) the visual-memorization shortcut, where VLMs rely on memorized visual patterns; (2) the knowledge-recall shortcut, where models leverage memorized factual knowledge instead of interpreting the diagram; and (3) the Clever-Hans shortcut, where models exploit superficial language patterns or priors without true comprehension. We evaluate 15 open-source VLMs from 7 model families on Chimera and find that their seemingly strong performance largely stems from shortcut behaviors: visual-memorization shortcuts have slight impact, knowledge-recall shortcuts play a moderate role, and Clever-Hans shortcuts contribute significantly. These findings expose critical limitations in current VLMs and underscore the need for more robust evaluation protocols that benchmark genuine comprehension of complex visual inputs (e.g., diagrams) rather than question-answering shortcuts.

Reward design remains a critical bottleneck in visual reinforcement learning (RL) for robotic manipulation. In simulated environments, rewards are conventionally designed based on the distance to a target position. However, such precise positional information is often unavailable in real-world visual settings due to sensory and perceptual limitations. In this study, we propose a method that implicitly infers spatial distances through keypoints extracted from images. Building on this, we introduce Reward Learning with Anticipation Model (ReLAM), a novel framework that automatically generates dense, structured rewards from action-free video demonstrations. ReLAM first learns an anticipation model that serves as a planner and proposes intermediate keypoint-based subgoals on the optimal path to the final goal, creating a structured learning curriculum directly aligned with the task's geometric objectives. Based on the anticipated subgoals, a continuous reward signal is provided to train a low-level, goal-conditioned policy under the hierarchical reinforcement learning (HRL) framework with provable sub-optimality bound. Extensive experiments on complex, long-horizon manipulation tasks show that ReLAM significantly accelerates learning and achieves superior performance compared to state-of-the-art methods.

Large language models (LLMs) are known to generate politically biased text, yet how such biases arise remains unclear. A crucial step toward answering this question is the analysis of training data, whose political content remains largely underexplored in current LLM research. To address this gap, we present in this paper an analysis of the pre- and post-training corpora of OLMO2, the largest fully open-source model released together with its complete dataset. From these corpora, we draw large random samples, automatically annotate documents for political orientation, and analyze their source domains and content. We then assess how political content in the training data correlates with models' stance on specific policy issues. Our analysis shows that left-leaning documents predominate across datasets, with pre-training corpora containing significantly more politically engaged content than post-training data. We also find that left- and right-leaning documents frame similar topics through distinct values and sources of legitimacy. Finally, the predominant stance in the training data strongly correlates with models' political biases when evaluated on policy issues. These findings underscore the need to integrate political content analysis into future data curation pipelines as well as in-depth documentation of filtering strategies for transparency.

Large language models (LLMs) excel at operating at scale by leveraging social media and various data crawled from the web. Whereas existing corpora are diverse, their frequent lack of long-term temporal structure may however limit an LLM's ability to contextualize semantic and normative evolution of language and to capture diachronic variation. To support analysis and training for the latter, we introduce CHRONOBERG, a temporally structured corpus of English book texts spanning 250 years, curated from Project Gutenberg and enriched with a variety of temporal annotations. First, the edited nature of books enables us to quantify lexical semantic change through time-sensitive Valence-Arousal-Dominance (VAD) analysis and to construct historically calibrated affective lexicons to support temporally grounded interpretation. With the lexicons at hand, we demonstrate a need for modern LLM-based tools to better situate their detection of discriminatory language and contextualization of sentiment across various time-periods. In fact, we show how language models trained sequentially on CHRONOBERG struggle to encode diachronic shifts in meaning, emphasizing the need for temporally aware training and evaluation pipelines, and positioning CHRONOBERG as a scalable resource for the study of linguistic change and temporal generalization. Disclaimer: This paper includes language and display of samples that could be offensive to readers. Open Access: Chronoberg is available publicly on HuggingFace at ( https://huggingface.co/datasets/spaul25/Chronoberg). Code is available at (https://github.com/paulsubarna/Chronoberg).

Collaborative robots (cobots) are a core technology of Industry 4.0. Industry 4.0 uses cyber-physical systems, IoT and smart automation to improve efficiency and data-driven decision-making. Cobots, as cyber-physical systems, enable the introduction of lightweight automation to smaller companies through their flexibility, low cost and ability to work alongside humans, while keeping humans and their skills in the loop. Industry 5.0, the evolution of Industry 4.0, places the worker at the centre of its principles: The physical and mental well-being of the worker is the main goal of new technology design, not just productivity, efficiency and safety standards. Within this concept, human trust in cobots and human autonomy are important. While trust is essential for effective and smooth interaction, the workers' perception of autonomy is key to intrinsic motivation and overall well-being. As failures are an inevitable part of technological systems, this study aims to answer the question of how system failures affect trust in cobots as well as human autonomy, and how they can be recovered afterwards. Therefore, a VR experiment (n = 39) was set up to investigate the influence of a cobot failure and its severity on human autonomy and trust in the cobot. Furthermore, the influence of transparent communication about the failure and next steps was investigated. The results show that both trust and autonomy suffer after cobot failures, with the severity of the failure having a stronger negative impact on trust, but not on autonomy. Both trust and autonomy can be partially restored by transparent communication.

Preschool children with language vulnerabilities -- such as developmental language disorders or immigration related language challenges -- often require support to strengthen their expressive language skills. Based on the principle of implicit learning, speech-language therapists (SLTs) typically embed target morphological structures (e.g., third person -s) into everyday interactions or game-based learning activities. Educators are recommended by SLTs to do the same. This approach demands precise linguistic knowledge and real-time production of various morphological forms (e.g., "Daddy wears these when he drives to work"). The task becomes even more demanding when educators or parent also must keep children engaged and manage turn-taking in a game-based activity. In the TalBot project our multiprofessional team have developed an application in which the Furhat conversational robot plays the word retrieval game "Alias" with children to improve language skills. Our application currently employs a large language model (LLM) to manage gameplay, dialogue, affective responses, and turn-taking. Our next step is to further leverage the capacity of LLMs so the robot can generate and deliver specific morphological targets during the game. We hypothesize that a robot could outperform humans at this task. Novel aspects of this approach are that the robot could ultimately serve as a model and tutor for both children and professionals and that using LLM capabilities in this context would support basic communication needs for children with language vulnerabilities. Our long-term goal is to create a robust LLM-based Robot-Assisted Language Learning intervention capable of teaching a variety of morphological structures across different languages.

Human autonomy and sense of agency are increasingly recognised as critical for user well-being, motivation, and the ethical deployment of robots in human-robot interaction (HRI). Given the rapid development of artificial intelligence, robot capabilities and their potential to function as colleagues and companions are growing. This systematic literature review synthesises 22 empirical studies selected from an initial pool of 728 articles published between 2011 and 2024. Articles were retrieved from major scientific databases and identified based on empirical focus and conceptual relevance, namely, how to preserve and promote human autonomy and sense of agency in HRI. Derived through thematic synthesis, five clusters of potentially influential factors are revealed: robot adaptiveness, communication style, anthropomorphism, presence of a robot and individual differences. Measured through psychometric scales or the intentional binding paradigm, perceptions of autonomy and agency varied across industrial, educational, healthcare, care, and hospitality settings. The review underscores the theoretical differences between both concepts, but their yet entangled use in HRI. Despite increasing interest, the current body of empirical evidence remains limited and fragmented, underscoring the necessity for standardised definitions, more robust operationalisations, and further exploratory and qualitative research. By identifying existing gaps and highlighting emerging trends, this review contributes to the development of human-centered, autonomy-supportive robot design strategies that uphold ethical and psychological principles, ultimately supporting well-being in human-robot interaction.

Transferring reasoning capabilities from larger language models to smaller ones through supervised fine-tuning often fails counterintuitively, with performance degrading despite access to high-quality teacher demonstrations. We identify that this failure stems from distributional misalignment: reasoning traces from larger models contain tokens that are low probability under the student's distribution, exceeding the internal representation capacity of smaller architectures and creating learning barriers rather than helpful guidance. We propose Reverse Speculative Decoding (RSD), a mechanism for generating student-friendly reasoning traces in which the teacher model proposes candidate tokens but the student model determines acceptance based on its own probability distributions, filtering low probability tokens. When applied to Qwen3-0.6B, direct distillation of s1K-1.1 reasoning trace data degrades average performance across major reasoning benchmarks by 20.5%, while the same model trained on RSD-generated reasoning traces achieves meaningful improvements of 4.9%. Our analysis reveals that low probability tokens constitute the critical bottleneck in reasoning ability transfer. However, cross-model experiments demonstrate that RSD traces are model-specific rather than universally applicable, indicating that distributional alignment must be tailored for each student architecture's unique internal representation. Left: Reasoning trace generation process where RSD produces student-friendly reasoning traces in which the teacher proposes candidate tokens, while the student accepts only those with high probability under its own distribution. Right: Average accuracy on major reasoning benchmarks (AIME24, AIME25, GPQA Diamond, and MA TH500) for (i) the base student model, (ii) a student trained on pre-existing high-quality reasoning traces (s1K-1.1), Recent advances in reasoning-focused language models have emerged through the strategic combination of reinforcement learning (RL) and supervised fine-tuning (SFT) (DeepSeek-AI et al., 2025). These two methods play distinct yet complementary roles in developing sophisticated reasoning.

In this paper, we offer a simple argument for the conclusion that the outputs of large language models (LLMs) are meaningless. Our argument is based on two key premises: (a) that certain kinds of intentions are needed in order for LLMs' outputs to have literal meanings, and (b) that LLMs cannot plausibly have the right kinds of intentions. We defend this argument from various types of responses, for example, the semantic externalist argument that deference can be assumed to take the place of intentions and the semantic internalist argument that meanings can be defined purely in terms of intrinsic relations between concepts, such as conceptual roles. We conclude the paper by discussing why, even if our argument is sound, the outputs of LLMs nevertheless seem meaningful and can be used to acquire true beliefs and even knowledge.